PROJECT SUMMARY Long spans of tandemly arrayed, highly repetitive sequences known as satellite repeats form an important and underexplored component of eukaryotic genomes. These satellite regions play vital roles especially in centromere and telomere function, and their dysregulation has been associated with several human genetic disorders. The repetitive nature of satellite sequences render them nearly impossible to assemble, thwarting efforts at further analysis. We have devised a pair of approaches that lets us explore patterns of kmer ?word? repeats and large-scale structures from unassembled reads to quantitatively describe the satellite repeat composition of a genome. We have discovered that these features are highly variable across lines, opening the door to understanding how satellite repeat variability evolves and how it impacts the biology of the organism. In Specific Aim 1 we will characterize the patterns and rates of turnover of satellite repeats in Drosophila. We will apply our novel computational methods to quantitatively characterize satellite repeats using DNA sequence data from population re-sequencing surveys and reference genomes of Drosophila species. Long- read sequences from PacBio runs and using GemCode technology from 10x Genomics will allow anchoring of a subset of repeats to neighboring euchromatin. These data will allow us to identify and quantitate satellite repeats, determine their species distribution, and infer rates of change and inter-satellite correlations within species and along the phylogeny. We apply Gaussian process models to recapitulate the rates and patterns of change of satellite repeats in fly genomes, and test whether patterns of satellite repeats are consistent with models of mutation-drift balance. In Specific Aim 2 we will determine mutational profiles of satellite DNA changes in mutation-accumulation lines of Drosophila. Satellite repeats display rapid turnover across closely related species, which motivates examining whether a purely mutational process is driving the evolutionary divergence. By contrasting satellites from genomes of flies derived from mutation accumulation lines, we will quantify key attributes of the mutational divergence. We will also construct reversible mutants of Su(var)3-9, which is required for heterochromatin maintenance, and test its effects as a sensitized background for mutation accumulation. In Specific Aim 3 we will test models of centromere- and telomere-drive by scoring segregation in F1s that have large differences in satellite repeat abundances. During female meiosis, any factor with an increased chance of ending up in an ovum instead of a polar body will have great evolutionary advantage. Prior publications suggest that such meiotic distorters play a role in shaping the evolution and distribution of satellite repeats. We will make F1 females in crosses between lines with specific differences in satellite repeats, and then score the progeny of these F1 females by sequencing to search genome-wide for distortions from Mendelian segregation. In sum, this project leverages novel analytical approaches to address long-standing and fundamental questions about satellite repeat evolution.